The Effects of Early Sexual Experiences: A review and Synthesis of Research

Larry L. Constantine

(from Children and Sex – Constantine & Martinson (eds) Little, Brown, Boston

Presented at the conference, Enfance et Sexualite, University of Quebec at Montreal, September 1979.

SCOPE OF THE STUDY

This study began in 1972 as an unstructured review of the literature on the effects of incest and other types of sexual encounters between adults and children. However, as questions about childhood sexuality and particularly the effects of early experiences became subjects of considerable debate and no small controversy, it was deemed appropriate to expand the project, to include sexual encounters between children, and complete it. This review is still by no means comprehensive; it is fact, selective in that it deals with a body of research and clinical literature concerned with the outcome of childhood sexual experiences, that is, with both the immediate and the ultimate impact of the experience. The primary criterion for inclusion was that a study either report differential outcome or contain material relevant to an attempt to count for differential outcome (single case reports are excluded). The chapter therefore ignores the sizable literature on offenders (the adult participants) as well as investigations on what might be termed the epidemiology of incest and sexual abuse. Many oft-cited sources were consulted but, because of their focus on treatment considerations or adult offenders to the exclusion of impact on the children, were not included in the tabulations for this study (e.g., Cormier et al., 1962) Reviews (Gagnon and Simon, 1970; Henderson, 1972) and opinion and speculative writings on the subject were examined but not tabulated.

In all, 30 studies of the impact of childhood incest and sexual encounters with adults were analyzed (Table 17 – 1). These studies cover a span of more than 40 years (although half are from the last 15 yea and include over 2500 subjects reporting early sexual experiences. With the exception of the Bender and Grugett (1952) follow-up of subjects originally reported by Bender and Blau (1937), there is no known overlap, although part of the Chaneles (1967) and Tormes (1968) studies may have drawn on the same population. Twelve sources (studies of clinical populations) are clinical case analyses or studies of subjects in psychotherapy; nine (studies of legal populations) deal with criminal cases brought to the attention of legal authorities; the remaining nine employed primarily nonclinical, noncriminal samples obtained in various ways (general population studies).

The outcome studies reviewed may be considered to be of three types: (1) research on children shortly after the experience occurred, (2) research specifically on adult adjustment of subjects known to have had such experiences as children, and (3) research involving adults who are incidentally discovered to have had such experiences. Only five studies are of the last type (Benward and Densen-Gerber, 1975; Finkelhor, Chapter 11 in this volume, 1978; Gagnon, 1965; James and Meyerding, 1977; Landis, 1956). Ten are of the first type, 13 of the second. One (Lukianowicz, 1972) has elements of all three.

The studies investigated populations ranging from two (Yorukoglu and Kemph, 1966) to 500 (Landis, 1956) and research methodology ranging from questionnaires and sexual autobiographies to case records analyses and in-depth interviews. The sophistication of design and care in reporting range from passable to excellent, Included in the compilation are four studies of general populations – Bernard (1979, and Chapter 15 in this volume), Finkelhor (Chapter 11), Nelson (Chapter 13), and Symonds et al. (Chapter 12) – published in this volume. Two of these (Symonds et al. and Nelson) employed a novel technique to obtain subjects: ads were placed in periodicals asking to hear from people with incest experiences.

The most sophisticated studies are also among the most recent. Meiselman (1978) not only integrates a thorough and critical review of the literature on incest with her own findings on 58 incest participants in psychotherapy, but also uses for comparison a control group of 100 subjects selected at random from among other patients of the same clinic. Tsai et al. (1979 and Chapter 16 in this volume) utilize two control groups, comparing sexual victims who sought therapy or counseling to both those who never sought such help and a control group of nonvictims. Their informants were solicited as in the Symonds and Nelson studies. Finkelhor bases his sophisticated data analysis on a survey of nearly 800 students at six New England colleges and universities.

The sexual experiences of interest in this chapter are both incestuous and nonincestuous encounters between children and their age peers and between children and adults or substantially older partners. Of the four categories of interest, only three are represented in the literature reviewed: incest with peers, incest with nonpeers, and other sexual encounters with nonpeers. The fourth category, nonincestuous experiences with other children, is not well represented in the outcome literature, although it has often been the focus of anthropological and ethnographic study. (See both Currier, 1979 and Chapter 2 in this volume, and Berger, Chapter 18 in this volume.)

For purposes of this chapter, studies of incestuous and nonincestuous encounters have been aggregated for two reasons. The first was a matter of practicality, in that twice as many outcome studies of incest have been made and, of the third of the studies under review that dealt primarily with nonincestuous experiences, over half also included some cases of incest. It was also a matter of theoretical importance, since there is an open empirical question whether incest is more or less damaging than sexual encounters with non–family members; this issue can only be addressed by including studies of both phenomena.

For similar reasons, no sharp upper limit on age of subjects was set. It is a matter of some debate, especially in the psychoanalytical literature, whether incest is more injurious prior to or after puberty. Certainly age warrants investigation as a possible factor in differential outcome. Thus, some studies tabulated deal with subjects whose experiences may have occurred in late adolescence, although prepubertal experiences account for the greatest proportion and the two largest studies deliberately limited their sample to these.

OUTCOMES

This chapter will make reference to negative, neutral, and positive outcomes. A negative outcome is a finding or conclusion that childhood incest or other sexual encounters were disturbing or traumatic or impaired the children who participated (at the time or later, as adults) in social, emotional, or intellectual functioning. All the studies reviewed but one report some negative outcomes. In some instances negative outcomes are inferred from the nature of the study: the two studies

Table 17-1. Outcome Research on Childhood Sexual Encounters

a + indicates study also included a control population

b Indicates reported outcomes, from very negative (– –) through neutral (0) to positive (+). First symbol indicates most frequent outcome reported.

finding high incidence of incest in the backgrounds of drug treatment patients (Benward and Densen-Gerber, 1975) and of prostitutes (James and Meyerding, 1977) are assumed to represent only negative finding since the authors are arguing, justifiably or not, for a causal lie between the childhood experiences and their subjects’ later social circumstances.

A neutral outcome is a report that some (or all) subjects were not harmed or damaged by their experience or did not differ significantly from control populations or the population at large. Some studies may report neutral outcomes even though subjects exhibit some impairment if it can be shown or argued that the childhood sexual experience was neither causal nor contributory; others report neutral outcomes only in the absence of all symptoms of damage.

A study is considered to report some positive outcomes if benefits are claimed to have arisen from the experience, for some subjects, or if the experience was evaluated positively by the subjects.

The evaluation and interpretation of long-term effects pose numerous methodological problems. Certainly physical beatings must be considered harmful, yet some abused children do not seem to suffer ill effects as adults. The absence of adult symptoms traced to early sexual experience cannot be taken simply as evidence for the harmlessness of those experiences. Conversely, psychosocial impairment in adults which were sexually victimized as children cannot be said to demonstrate the dangers of those encounters, for the impairment observed may originate in other experiences or chronic stressors or even be, in some measure, reflections of constitutional factors. Only well-chosen control groups can resolve these causal ambiguities. Unfortunately, controlled studies are rare in this field.

It can be argued that the great variation in criteria and in methodological soundness among the studies reviewed precludes comparison and generalization. However, as it is differential outcome that is the focus, comparisons within each study can still be made. Moreover, differentiating factors that hold across studies of widely varying methodology and sophistication are thereby lent greater credence.

Harm

Are incest and sexual encounters with older persons harmful to children? How severe are the effects, and are they of brief or lasting duration? The answers to these questions vary widely from study to study.

Landis (1956), drawing on 500 college students who, prior to puberty, had sexual experiences with adults, reported that only 3 percent were permanently damaged and no harm, permanent or temporary,

resulted for 81 percent of males and 66 percent of females. Gagnon (1965) found a similar statistical breakdown in 333 women from the Kinsey studies who had reported prepubertal sexual experiences with adults; only 5 percent (18) were severely impaired for any reason, and only three of those 18 attributed their problems to their early sexual experiences. Three-quarters of the total sample had no apparent adult maladjustments. Because the Gagnon sample includes a disproportionate number of college-educated subjects, like the Landis sample it is biased in the direction of including the more socially and psychologically adaptive cases.

Clinical and court cases are, of course, biased in the other direction. Thus, among a group of 58 psychotherapy clients studied by Meiselman (1978), 37 percent of those under 12 years of age at the time of incest experience and 17 percent of those who were 12 or over were judged to be seriously disturbed. However, overall they did not differ markedly from other psychotherapy clients at the same clinic, being slightly more disturbed and presenting more problems, principally of a sexual nature.

In contrast, Brunold (1964), following up on the adult lives of 62 victims of sexual assault, found only one who claimed her "whole life" was "ruined," but all others denied lasting damage. No lasting effects were discernible in 90 percent of the cases. And Lukianowicz (1972), reporting on 55 cases of incest among 1350 unselected admissions to hospitals and a home for girls, states that six of 26 daughters sexually involved with their fathers had no apparent ill effects. Except for one patient with a reactive depression triggered by humiliation and brutality at the hands of her brother, no other participants in other forms of incest showed "any morbid consequences or ill effects of their incestuous behavior." Rasmussen (1934) reported that 46 of 54 child victims were, as adults, none the worse in terms of mental health and social adjustment.

At one extreme are studies such as those of Benward and Densen-Gerber (1975), Molnar and Cameron (1975), and Peters (1976), which report only negative findings; at the other extreme stands Ingram (1979, and Chapter 14 in this volume), who reports there was no evidence that any of the boys in his sample was worse off for his sexual activity with an adult, and that many may have benefited from relationships with loving adults outside of their affectionately deprived homes.

The variability in researchers’ findings and conclusions does not prevent generalization, however. At least one elementary conclusion, that some children are not detrimentally affected by early sexual experiences with adults or with family members, can be asserted with confidence. Of the 30 studies reviewed, 20 report at least some subjects without ill effects; 13 of those conclude that, for the majority of subjects, there is essentially no harm; and six even identify some subjects for whom, by self-evaluation or other criteria, the childhood sexual encounter was a positive or possibly beneficial experience.

It is worth noting that the basic finding of absence of negative effects in some subjects is not limited to surveys or nonclinical studies but also shows up whether the sample is from legal or clinical sources (a general population, whether the measures employed involve self-assessment, social and behavioral variables, or in-depth interviews in psychotherapy; and whether short-term or long-term effects are scrutinized. In some cases there is room for interpretation not always intended by the author. For example, Browning and Boatman (1977), while claiming no neutral outcomes, do indicate complete absence of all symptoms in the six-month followup of one girl who had been involved in father-daughter incest and for whom no therapy was undertaken.

It is a challenge, however, to account for the variance in extent and severity of reported harm from childhood sexual experiences. Not surprisingly, the nature of a study had a great deal to do with the nature of the reported outcome. Clinical case studies and others dealing with people seeking or referred for therapy would be expected to report more negative outcomes than those surveying a general nonclinical population, and when the sample is taken from court referrals and cases of sexual assault, the outcomes also can be expected to be more negative. Such is the case in the studies reviewed for this chapter: of eight studies of childhood sexual experiences among the general population or with nonselected, nonclinical samples, half reported some subjects with positive experiences and half reported at least some which were unaffected. None of these studies found only negative results. By contrast, six of nine studies based on legal sources reported only negative results and although eight of thirteen clinical studies reported some subjects unharmed, only one reported some positive results (Table 17 – 2).

Whether short-term or long-term outcomes were investigated also influenced the findings. Five of eight studies based wholly on short term effects reported only negative outcomes, while 15 of 18 studies based in whole or in part on long-term effects report some neutral or positive outcomes (Table 17 – 3). One might surmise that the short-term studies were looking more at children’s surface reactions – often characterized by anxiety and guilt – than at the deeper impact of the experience. Alternatively, it might be argued that over the long term any trauma will have faded, and the lasting effects of even highly negative

Table 17-2. Number of Studies Reporting Various Outcomes by Type of Sample

Table 17-3. 
Number of Studies Reporting Various Outcomes by Duration of Effects Studied

experiences may be minimal. Closer scrutiny of the studies reveals evidence to support both these explanations; however, the latter argument is insufficient in that some studies report neutral or positive effects over the short as well as the long term.

Small matters, too, may introduce major sampling biases that could profoundly influence findings in a study. Thus, when Ms. magazine runs a feature on the horrors of incest and asks readers to tell their stories, they are flooded with letters, all recounting anxiety, guilt, and permanent harm. But when Nelson (Chapter 13 in this volume) advertised for persons who had incest experiences, "good or bad," only 39 percent of experiences reported to her were described as negative, and fewer than 30 percent of the respondents felt they were harmed.

Reactions

Children's immediate reactions to incest and other sexual encounters are an interesting aspect of outcome often neglected in the clinical literature. Most of the studies directly inquiring into reactions have been nonclinical. Landis (1956) found that 39 percent of males and 17 percent of females had reacted with interest or surprise but were not frightened, while the remainder responded that they had been frightened, shocked, or emotionally upset. Gagnon (1965) reports that the "overwhelming majority" of his subjects reacted negatively, most often with fright, but some did respond with mixed curiosity or with indifference. In the Nelson study the four most frequently checked adjectives were all positive; fear was only the fifth most frequent checked. The siblings involved in incest in the study by Symonds et al. generally reported positive feelings; although some felt guilt at first, most felt closer to the sibling and more trusting as a result. Tsai et al. report that their clinical group reported more upset, more pain, and more negative feelings than did the nonclinical group, indicating that immediate reactions to the situation are correlated with outcome, a conclusion supported also by comparisons among studies.

But the most frequently cited reactions in analysis of the entire group of 30 studies are anxiety (or fear) and guilt. Ten studies found anxiety or fear to be a common reaction, and nearly half cited guilt; the characteristic reaction, one (Sloane and Karpinski, 1942) considering it to be the most outstanding and universal feature of incest outcome.

Guilt 

Researchers are particularly articulate on the matter of guilt and direct quotations are in order. Weiner (1962) concludes that guilt does "not relate to the incest itself but to the disruption of the home after discovery." Kaufman et al. (1954) concur that not the sex itself but rather the disruption of the home generates extreme guilt and anxiety.

Molnar and Cameron (1975) say of one case ("Allen") that, if the victim does not realize there is a taboo, the effects should be nil. This parallels a case (number 2) in Weiner (1962), in which no guilt was manifested in a daughter who, from an early age, grew up in a highly sexualized endogamous family milieu. The same point is made from the other way around by Finch (1973) who states: "At the time of seduction, most children have learned that sex is not an acceptable activity, so guilt usually follows any sexual involvement."

In a widely cited paper, Bender and Grugett (1952) say: "At first the children showed no guilt, but this tended to develop as they were separated from their sex object and means of gratification, and as they were exposed to the opinion of parents and court officials. It occurred especially with the more intelligent children and seemed in part a reflection of adult censure." And in the earlier study, Bender and Blau (1937) say of "Frances" that she "did not present any neurotic or emotional reaction to the situation,... [but] merely agreed that it must have been ‘bad’ because the judge had told her so." The same study tells of another child who perhaps should be commended for being ahead of his time. In light of today's changing sexual values, the case is ironic: "Although he had been told that the purpose of sex was procreation, he refused to believe it and thought that it was only for pleasure." Symonds et al. report that 29 of 109 subjects experienced guilt attributed to the impact of societal values. The Tsai et al. study provides differential information on guilt: clinical cases evidenced more guilt about secrecy and concerning the sexual activities themselves than did nonclinical informants.

Anxiety And Fear 

Differential outcomes concerning anxiety and fear are less clear than for guilt; many writers lump all three together. Landis (1956) found that approximately half of those in his study with negative immediate reactions told parents about the experience, but only a third of those who were "surprised" but not frightened and only 12 percent of those who reacted "with interest" told parents. Thus professionals, especially police authorities, are unlikely to hear of neutral or positive reactions.

Benward and Densen-Gerber (1975) found the reaction to be dependent on the type of sexual encounter. Those between age peers were mostly matters of active participation or what they termed passive consent, and the reactions were characterized by ambivalence in response to stressful situations and fear of discovery. Across generations, however, nearly half were forced sexual relations, which triggered fright and confusion.

Anxiety is most often reported as absent in nonclinical, nonlegal cases or in clinical or legal cases under special circumstances. Lukinowicz (1972), for example, attributes the absence of anxiety and guilt to an accepting family and social subculture that viewed incest matter-of-factly or as normal.

Many reviews and overviews on incest and child "molestation" state that the reactions of parents and other adults can be more frightening to children than the sexual experiences themselves. Eight studies found this to be a factor, but in the Landis study (1956) only 5 percent said they had been more upset by their parents' reactions, and Brunold mentions "some" for whom this was true.

Self-evaluation 

Some recent studies report subjects' overall evaluations of their experiences. Of those responding to the ads of Symonds et al., 80 percent felt positively about their experiences toward their incest partner, but a greater portion of sibling partners than offspring involved in parental incest had positive experiences and, as a group, those whose experiences included humiliation and so-called discipline had the most negative reactions.

The respondents in the Symonds study were predominantly male, which makes it difficult to compare responses by sex. This was not a problem in the nearly balanced study by Nelson in which just over half of all experiences were evaluated as positive, but only a quarter of the experiences of females were said to have been positive. In adult-child relationships, nearly all evaluations by the older partners were positive, while younger partners evaluated a quarter of theirs positive. On the other hand, exactly half of the experiences of the younger partners in peer relationships were reported as being positive. As in other studies, younger partners were more often female than male. But among males only, 87 percent of the experiences of those who were the older partners were reported as positive, while only 62 percent were so described when the respondent was the younger.

Nelson also found a general relationship between age and evaluation of experiences. Only 18 percent of experiences occurring prior age 10 were considered positive, rising to 40 percent at ages through 12, to 60 percent in early teens, and reaching 85 percent for teens 16 or older. The last is the only group among whom the proportion of experiences reported as positive by females exceeds that of males; no experience in late adolescence was evaluated as negative by a female respondent.

Finkelhor, reporting in Chapter 11 on sibling incest among college students in a New England survey, found that 35 percent of females reported negative experiences, whereas only 22 percent of males did. Over half of the cases where the partner was much older were evaluated as negative, and 64 percent were evaluated as negative where force or coercion was employed.

The regression analysis of these data conducted by Finkelhor is particularly illuminating. Only three variables had significant correlations with evaluation. Age difference between partners was the most salient. The greater the number of years younger the respondent was, the more likely he or she was to perceive the incest as negative. The use of force or threats was also significantly related. (The third variable, nature of the incident, is discussed later in this chapter.)

Of considerable interest are factors in Finkelhor’s study found not have influenced evaluation. Age at the time of the experience, duration of the experience, sex of the respondent, whether the relations' was homosexual or heterosexual, and whether or not intercourse occurred seemed not to influence subjects' evaluations at all. Finkelhor concluded that girls' experiences were more negative because more them encountered coercion and had much older partners.

Meiselman (1978) also supports these findings. A greater number incestuous sisters than daughters reported sexual pleasure, and among sisters there was a close relationship between pleasure and voluntary participation with a brother close in age. When the age difference exceeded several years, the relationship was more likely to be exploitive, primarily serving the brother’s needs, and was negatively evaluated by the sister. Both Finkelhor and Meiselman cite the use of persuasion, bribes, and other power manipulations on the part of the older sibling, their use more frequently being associated with larger age differences.

Sexualization

Many studies report that children, even prepubertal children, become sexually awakened by childhood sexual encounters and may become quite sexual in their behavior. Ten of the studies reviewed mention this as common or typical. In an appropriate setting accepting of children's sexuality, such as in some of the communal families studied by Johnston and Deisher (1973) or in the group – marriagelike incestuous families described by Symonds, this could be an acceptable, even positive response. But, in more conventional settings, the conflict with social norms can be considerable. Bender and Blau (1937) reported on the valiant efforts of "Frank," an otherwise normal child introduced to sex an early age, who strove to repress his sexuality and develop new interests. They wrote of three children whose main difficulty was "intellectual bewilderment... precipitated by an inability to resolve the conflict between... stimulation of genital sexuality... and the social tendencies for sublimation in school and play." It is not an intrapsychic conflict about which they write but a struggle with adult social expectations.

The social consequences of early sexualization are potentially substantial. Chaneles (1967) says that "the incidence of such behavior suggests that those who have been victimized [sic] may become, as a result, much less puritanical in regard to general heterosexual norms." Brunold (1964) questions "whether the heightened sexual sensibility can be interpreted as injurious, provided that this does not lead to promiscuous sexual behavior." Sloane and Karpinski (1974) also argue for the possible positive aspects of even so-called promiscuous sexual behavior, crediting acting out as a substitute for neurotic symptom formation in the majority of their cases. But promiscuity is a matter of varied definition and tends to be applied to females more than males even for comparable behavior. Older studies (e.g., Weiss, et al., 1955) find it more often than do later works. Meiselman (1978) identified periods of promiscuous behavior in 22 percent of incestuous daughters but in 71 percent of sisters involved in incest. Her criteria for identifying promiscuity were "sex on a very casual basis" or changing sexual partners "very frequently," that is, "every few weeks or months." Weinberg (1955) also reports more promiscuity among incestuous sisters than among daughters. Gligor (cited in Meiselman, 1978) found only two of 57 incestuous daughters to be promiscuous, meaning involved in sexual activity with more than one partner in one period or indiscriminately acting out sexual impulses. It can be argued that even these criteria suggest not something pathological or even deviant but sexual behavior well within contemporary norms.

In comparison with a control group from the same sample who did not have prepubertal sexual encounters, Landis (1956) found that females with experience of whatever type were more likely to be nonvirgins and more likely to have experienced orgasm before marriage. Males with prepubertal experiences also were more likely to be nonvirgins and had started dating at a younger age than controls. The form of participation was significant for females: only 12 percent of controls and 20 percent of those with "accidental" encounters, but 31 percent of possibly cooperative "participants," said they had ever experienced orgasm. Although Gagnon (1965) reported no differences in orgasm rates, 73 percent of "collaborative" but only 57 percent of "accidental" victims had married. In all categories there was more than a usual amount of prepubertal sex play.

Once again, the analyses by Finkelhor are particularly useful. Regardless of.the nature of the experience, women in his study who had been involved in sibling incest were more sexually active than both those reporting no childhood sexual experience and those reporting only other (nonincestuous) types of experience. The quality of the experience (whether it was with a peer close in age and/or was evaluate positively) make some difference only when the experience occurred before age 9.

Women whose sexual relations with a sibling had been positive who had had relations with a peer had significantly elevated sexual self-esteem scores; if the experiences were negative and not with an age peer, self-esteem was lower than that of those with no early sexual experience. Positive peer experiences had more impact on self-esteem they occurred after age 9, and negative and nonpeer experiences ha more impact if they occurred earlier. Older girls with negative experiences became more sexually active but did not have higher self-esteem.

Finkelhor seems to make a valid conclusion that a positive, none exploitive sexual experience in adolescence with a peer well known to the girl could provide an early model for success in fusing sex and friendship, an important developmental task in adolescence. He is justifiably cautious in generalizing from these findings: much (about one-fourth) of sibling incest appears to be exploitive.

In general, the studies reviewed suggest that the subsequent sexualized behavior of children who participate in incest or sexual encounters with adults is not intrinsically a problem for the child but rather for parents and other adults. Sexual pursuits need not, as the anthropological literature attests, in any way interfere with other usual pursuits of childhood.

Before leaving this question, the matter of homosexual versus heterosexual experiences needs to be addressed. In general, the literature is equivocal on this point. While a few studies (e.g., Finkelhor) do find adult homosexual activities to be more common in those who have had childhood homosexual incest experiences or experiences with adults, these can be interpreted as representing early manifestation of sexual orientation as readily as they can be read as examples of the deviating effect of early experiences. The vast majority of children with early homosexual experiences of any kind do not become homosexual adults. Magnon (1965) did find a possible pattern of childhood homosexual play as a reaction to heterosexual experiences with adults, but this orientation did not persist.

Initiation and Participation

Do children ever actively initiate, or willingly participate in, incest or adult-child sexual relations? The stories of child and adult participant often conflict, especially in criminal cases where the stakes may be high (Cormier, et al., 1962; Gebhard et al, 1965). Reports by adults describing sexually seductive children must be read with caution and in context. Still, the research literature and clinical reports are in sufficient accord that there remains little doubt that some children do initiate sexual encounters and that many participate willingly even when they are the younger partners. Unequivocal findings on collaboration and/or mutual instigation are reported by nine studies reviewed and suggested in others. In some cases the relationship became known because a minor was soliciting sex with adults; in others the child's own statement of initiation or cooperation is available. In the Gagnon (196l) study, for example, 26 of 31 nonaccidental cases were provoked by the child, according to the child’s own statement, or were matters of mutual desire. Meiselman (1978), however, found outright seduction and active cooperation to be rare in girls involved in incest with fathers. Girls who had sex with siblings more often said their relationships began with mutual interest and participation.

Various writers, among them those presenting the least alarming pictures of childhood sexual experiences, point to the possibility for different intent and attributions on the part of older and younger participants. Ingram found the majority of boys in his sample to have been seeking affection and attention, but not necessarily sexual affection and attention. Even patently sexual interests may be different for the child participant, as shown by the fact that most of these boys preferred being masturbated over masturbating their adult partners.

Collaborators or active participants are the more likely (1) to experience pleasure; (2) as young adults, to be sexually active and (for males) to have experienced orgasm; (3) as children, to have engaged sex play; and (4) to evaluate their experiences positively. Gagnon (1965) reports that although all coerced encounters in his study were evaluated as negative and only 1 percent of accidental experiences were positive (84 percent were negative), 50 percent of collaborative experiences were reported as positive.

It should be noted that the previously cited heightened sexuality in children who were involved in childhood sexual encounters may have predated the experiences rather than having followed and, presumably, resulted from them. Meiselman (1978) argues that this is generally not the case, but Gagnon, intriguingly, found that above-average frequency of heterosexual sex play as generally reported prior to any sexual encounter with adults, but that homosexual play (reported in half of accidental victims but in two-thirds of multiple-incident "accidental" victims and collaborators) was only above average after the encounter with adults. This finding supports the notion of sexually interested and active children as collaborators in experiences with adults, but also would seem to suggest a possible reaction against these experiences, which were heterosexual.

Power

The issue of children's choice or consent in incest and sexual encounters with adults and the related matter of power are obviously related to the issues of cooperation and initiation. Power is a subtle but important element of interpersonal relations, with many facets, including not only the power wielded by adults, but also the power of the child to resist and to participate.

Force and coercion 

Most of the reported experiences, 90 percent of the incest cases examined by Justice and Justice (1979), for example, do not involve any actual physical force. Of the 64 emergency room cases studied by Peters (1976), 54 percent involved no actual force, although in 31 percent of cases verbal threats were used, and in 25 percent "tempting" was employed. Tormes (1968) says of the consistently negative experiences reported in his study that all victims had submitted because of personal threats.

Although some case reports claim to the contrary that guilt is reduced in cases where a child was forced, the evidence is generally consistent that the presence of force and/or coercion is associated with the more negative outcomes and its absence with the more positive ones. In the study by Symonds et al. those who fared worst were those subjected to brutality, humiliation, and so-called discipline. The majority of the incestuous sisters in the Meiselman study could be described as slightly negative about their experience, and one expressed revulsion more typical of daughters in that study; she had been threatened and humiliated by two older brothers. Finkelhor found force or threat force to be one of only three factors significantly correlated with evaluation of the experience. In the study by Landis, only a third of all females evidenced temporary or permanent damage, but 80 percent of victims of attempted rape did.

Meiselman, in part, credits gentleness and the absence of threat and intimidation for some better outcomes in incest between nieces and uncles and granddaughters and grandfathers. Peters (1976) at first suggests that "perhaps child victims react with less perceptible emotional response because the incidents were less severe, less brutal, and far less life-threatening than those involving adults or adolescents." He later concludes however, "that the child is unable to express her reaction and may retreat into emotional withdrawal which is then misinterpreted." Other findings suggest the first interpretation to be more salient.

Age

Age is related not only to ability but also to interpersonal attributions of power. It is interesting that in the Landis and in the Tsai et al. studies, attempted, but not actual, intercourse, which implies unsuccessful coercion, was associated with negative reactions. And the clinical group in the Tsai study had been subject to more pressure to comply and to keep the encounter secret than had the nonclinic group.

The findings concerning age at experience are inconclusive. Only five studies conclude that age does influence outcome: three conclude that older children are better off, two the opposite. Justice and Justice (1979) conclude that incest "appears to be least damaging psychologically to the younger child, provided she or he is not physically assaulted and traumatized." This would seem to support Peters' (1976) initial conclusion as well as other findings on the role of physical force in outcome. Further, Justice and Justice assert that "the younger child usually does not realize the significance of the sexual behavior and so does not suffer so much guilt," thus concurring with other findings on the origin of guilt. But Meiselman found just the opposite: more seriously disturbed patients among those whose incest experiences began before age 12 than after. And Nelson found more evaluations to be positive with increasing age at experience.

Finkelhor, controlling for age difference and force or threat, found that age was not a factor in evaluation of the experience, but early non-peer incest was associated with lowered self-esteem. Brunold (1964) and Gagnon (1965) also found age at experience not to be a factor outcome. The work of Tsai et al., the best controlled study to date, found that age at last sexual experience with an adult differentiated incest participants who had been in therapy from those who had not: the clinical group’s last experience came later, in adolescence, and had been of longer duration. Since the average ages at first experience did not differ between groups, duration appears to be more salient than age.

Age difference, on the other hand, may be very salient, although only two studies provide direct data concerning it. In the Finkelhor study age differences had the highest correlation of any factor related to outcome. Meiselman (1978) also refers to the relevance of age differences in reporting on sibling incest. Her most unequivocally positive evaluation is by one sister whose relations of several years with a young brother were intensely pleasurable and were described by her as the best sexual experiences of her life.

In other studies, the comparison is made between peer relations and relations with adults or between siblings and other, primarily father-daughter, incest. Benward and Densen-Gerber (1975) report that cross-generational incest experiences were more short-lived than those among generational peers, were more often experienced as forced or coerced, and resulted in more negative reactions. Of cases of brother-sister incest, Lukianowicz (1972) concluded that there are few if any problems because "there is no scolding, no threats, no punishment, and so these children do not later develop feelings of guilt and [do] later find it easy to substitute for their siblings new sexual partners from outside the family." Others have mentioned that siblings do not generally need to fear even if detected, the breakup of the home, a major source of guilt and anxiety for children sexually involved with parents.

Of eight studies furnishing support for the salience of a peer relationship in more positive outcomes, only Weinberg (1955) argues the reverse, stating that "daughters as a category of participants, were the best adjusted sexually and socially" being "very attached to their husbands" and "devoted mothers." Since more of them were married and fewer were described as promiscuous than was the case among sister participants, this dissenting opinion is plausible.

The findings concerning the effects on adolescent participants of incest or sex with adults are ambiguous and contradictory. However, clinical case studies and the reports of Tsai et al. and of Finkelhor suggest that heightened awareness of social norms and needs to conform during adolescence are likely to be associated with greater guilt.

Choice and Consent

In all, 14 of the sources reviewed report material indicating differential outcome as a function of the presence of force, coercion, or nonvoluntary participation. Meiselman (1978) provides some of the richest material on this issue. She found complaints of sexual problems to be characteristic of patients who had participated in incest, including 8 percent of incestuous daughters. Only two women in her study did not evidence sexual problems, one who had successfully resisted her father’s first advance and another whose father desisted when she asked him to. These two were described as "competent and assertive as adults.... As children, they had insisted, without any help from their mothers, that the incest relationship be ended, and as adults they refused to tolerate abuse from boyfriends or husbands." Two others had some sexual problems that they had overcome prior to therapy: one had ended the incest on her own, the other with the help of a supportive mother. Among incestuous sisters were two with only mild sexual problems in the context of generally good adjustment: "Like the daughters who were minimally affected by incest, these sisters had been well-adjusted children from relatively healthy, nonabusive families. Even more significantly, their incestuous involvement had arisen from their own sexual curiosity and had not been imposed on them by the brother." Both were extremely competent. They were the only sisters not labeled as masochistic by their therapists and were experimental but not compulsive in the sexual relations. It appears from this analysis that a child’s perception of control, the ability to choose or reject participation in incest or sex with adults or siblings, is an essential ingredient in a healthy outcome from the experience, and this matter is as often decided by the ability to end an undesired relationship as it is by active entry into a desired one. For example, the best outcome reported by Sloane and Karpinski (1942) involves the case of "Jean" who, at 19 successful terminated a sexual relationship with her father. Gagnon (1965), and Ingram, Symonds et al., and Nelson in this volume also report findings supportive of this view.

It is a question of some import whether or not a particular child of a certain age in a given circumstance can realistically be capable of consenting to a specific sexual activity. Informed consent requires not only knowledge but the ability to resist successfully if the choice is to demur. Of particular interest are cases of "passive acceptance," in which the child is actually an unwilling cooperator. Most studies lump these cases either with noncollaborative or with collaborative relationships; the few that do not have concluded that this group are the worst off of all, in terms of outcome. Benward and Densen-Gerber (1975) found those who gave passive (that is, actually unwilling) consent carried the greatest psychological burden and were often confused and humiliated by their experience. They felt guilty and ashamed and yet had also felt helpless to disengage.

Passive consent is more likely to occur in situations where a substantial age difference and differential role expectations make it difficult for the younger partner to decline. In the sibling relationships studied by Finkelhor, which he characterizes as including all the usual elements of manipulation, persuasion, and use of power typically attributed to sibling conflicts, it is worth noting that the correlation with outcome was higher for age difference than for use of force or threat. This would seem indirectly to support the view that passive consenters have more negative experiences than active cooperators or, possibly, even non-consenters.

Family

Thus far, this chapter has paid scant attention to a major preoccupation of the literature on incest (and also, to a somewhat lesser extent discussed in the literature on adult-child sexual encounters). Little purpose would be served by summarizing here the rich composite picture of family structure and family dynamics in these cases (Cavellin, 1966; Hersko, 1961; Kaufman et al., 1954; Lustig et al., 1966; Machotka et al., 1977). However, certain factors, environment and familiarity, emerge as affecting differential outcome significantly.

Environment 

Sixteen of the studies reviewed, the largest proportion for any factor considered, include substantive findings supporting the significance of the family environment in influencing outcome. The strongly negative outcomes involve cases of highly disturbed, multi-problem families; studies that report only or primarily negative outcomes tend to confirm this finding. Justice and Justice (1979) conclude that "it is not the sexual activity itself that is the problem but the kind of disturbed and troubled relationship in the family," and "the most lasting consequences came from the disturbed... relationships in the family." Meiselman (1978) reports that "the most immediate effect of incest was found to be the further disruption of a family that was disturbed to begin with."

At the other end are cases of incest or childhood sexual experiences in otherwise healthy families. The outcome in these cases is uniformly more positive. Bender and Blau (1937) refer to "Albert," a frank and amiable boy who evidenced no guilt or anxiety concerning his sexual relations with an older male. They describe his family as "poverty stricken but otherwise adequate." Nelson, whose sample had the second highest proportion of positive evaluations, found evidence of family breakdown in only 19 out of 100 situations. Ingram used strict objective criteria to judge whether a home environment was satisfactory or unsatisfactory; in all 14 of the families where one or both parents were judged to be satisfactory, it turned out that the boy had informed one or both about his sexual experiences with men. In the eight of these cases where the boy, subsequent to disclosure, was referred for counseling on that account, a single session sufficed. It should be emphasized that the vast majority of informants in all nonclinical, nonlegal studies had not told their parents of their experiences, and most had never told anyone; thus, this finding is further evidence of the accuracy of Ingram's assessment of these 14 families.

Six studies provide evidence that a particular aspect of the family environment is of special significance. When the mother is supportive and accepting, even otherwise negative experiences diminish in impact. For example, the only daughter in the Meiselman study (1978) who minimized the effects of her incest experience had a particular supportive mother who allowed the daughter to express her feeling freely and thereby helped her to "work it through."

Three studies (Lukianowicz, Weiner, and Yorukoglu and Kemp) also identified the presence of a distinctly tolerant and openly sexual family subculture as a factor in more positive outcomes. In addition, eight studies (Bernard, Brunold, Chaneles, Ingram, Landis, Peters, Tormes, and Tsai et al.) document the harmful contribution of adults who react highly negatively.

Familiarity 

It has now been well established that most "sexual assaults" on children, if not incestuous, at least involve persons known to the child. Although the theoretical literature is outspoken on the greater psychological damage that results from incest and sexual involvement of children with adults they know, the outcome literature offers scant support for or against the theories. Landis (1956) reports that it took longer for his respondents to recover when the adult was known to the child than when he was a stranger. Peters (1976) also claims that more severely negative consequences follow sexual assault by someone known to the victim. But 87 percent of the encounters with strangers in the Landis sample were cases of exhibition only which Finkelhor found to be the second largest of the three factors he considers significantly related to the subject’s evaluation of the experience. Finkelhor's comparisons of subjects reporting sibling incest versus nonincestuous experiences do not show either group to consistently fare better. The clinical and nonclinical groups of Tsai et al. did not significantly differ in percentage of experiences with relatives. Although Bender and Blau (1937) conclude that incest is more harmful than other forms of childhood sexual experiences with adults, it is not clear how their cases support this conclusion. Two other supporting sources (Chaneles, 1967; Tormes, 1968), apparently drawing on the same sample, compare only father-daughter incest with other unspecified cases of "sexual molestation," so they do not contribute to the analysis,

On a potentially related matter, the significance of the duration of a relationship on outcome, only four sources found that longer relationships were more harmful than others, the clearest finding being that of Tsai et al., that, as noted earlier, age at last experience differentiated clinical and nonclinical groups. Finkelhor’s analysis eliminated duration as a factor in itself. But recall that Benward and Densen-Gerber (1975) found a relationship between duration and peer relations as well as between consent and peer relations. Consensual relations (which tend to be more positive) in that sample were also likely to be shorter. Close scrutiny of case material suggests that childhood sexual relationships with adults last for one or two reasons: because they are mutual and experienced as positive or because the child is coerced by someone nearby. Thus, positive peer relationships and coerced nonpeer relationships should be those most likely to be continued.

CONCLUSIONS AND SYNTHESIS

To make sense of so large and diverse a body of findings as the 30 studies reviewed for this chapter, it would be helpful if a compact explanation would emerge, a parsimonious model to account for the greater portion of the consistent trends relating to differential outcome. Put simply, the research literature shows there to be no inbuilt or inevitable outcome or set of emotional reactions to incest or to sexual encounters of children with adults. The more negative outcomes are associated with ignorance of sexuality; with negative attitudes toward sex; with tense situations; with force, coercion, or brutality; and with unsupportive, uncommunicative, or judgmental adult reactions.

Briefly, the following are found to account for most of the variation in outcome and evaluation of the experience, ranging from negative to positive:

1. The child's perception of being free to participate or not

2. The child's prior knowledge and values concerning the sexual acts

3. The subsequent communications of family members and others about the experience

4. The quality of relations in the child's support system

A Model OF Outcome

Sexualization 

Because children are intrinsically sexual beings, actual sexual experiences will tend to sexualize their behavior and their construal of reality. Whether and to what extent "precocious sexuality" is problematic will depend on the social and familial values with which the child lives. Money’s (1973) threshold model provides the simplest and most general model for the process of awakening to sexual experience. The threshold for release of erotic response and actuation of sexual behavior is higher prior to puberty; therefore, less sexual activity will be evident in the child in the absence of direct, nonsymbolic stimuli.

Consent 

Consensual participation can take the form of active cooperation, passive (reluctant or unwilling) consent, and forced (or clearly coerced) participation. Sexual encounters of children who are cooperative are generally characterized by moderate to low anxiety, those involving passive consent by moderate to high anxiety, and those involving forced sexual experiences by high anxiety or fear.

Of overwhelming importance in immediate and long-term effects is the child’s perception of his or her participation as voluntary or involuntary. Other things being equal, the more mutual and voluntary the sexual experience is, as perceived by the child, the more positive the outcome will be. Passive consenters may be particularly vulnerable to negative outcomes because of the experience of being unwilling to participate but unable to resist.

Knowledge

The child's sexual knowledge must be recognized as consisting of two components, with different effects on the child's reactions and probably also on ultimate outcome. What will here be termed knowledge of moral negatives includes awareness of and sensitivity to taboos on sexual behavior and social role proscriptions, as well as general values attached to sex. (The negative sanctions and ascriptions are what is of relevance here.) Sexual knowledge refers to knowledge of human sexuality considered apart from cultural judgments. Individuals absorb the moral negatives about sex at different rates and in different ways; some become aware of but remain insensitive to them, but most conventionally reared children have probably learned that sex is a "no-no" by the time they start school.

The research reviewed for this chapter is generally consistent in showing the role of negative attributions, including that induced by later adult reactions, in elevating guilt in children about their sexual behavior. In extreme cases children have even been made to feel guilty

Table 17-4
 Guilt and Anxiety as a Function of Consent and Knowledge in Childhood Sexual Experiences

about sexual activities in which they were forced to participate. However, on the whole, guilt appears to be low in children who were forced participants, although not in cases of coerced consent. The significance of moral negatives in affecting outcome may be greatest during adolescence, although the research is inconclusive.

Little in the outcome research touches on the degree of sexual knowledge of the child participants, but since the "unknown" in general elevates anxiety, and it is known that children’s inability to label the genitals contributes to anxiety in connection with genital function, it is reasonable to assume a role for sexual knowledge in affecting outcome. The anthropological literature also would seem to support the conclusion that greater sexual knowledge reduces fear and anxiety about sexual activity except in situations involving force.

The interactive effects of knowledge and consent can be summarized as a 12-celled model (Table 17-4). Sexual ignorance alone is assumed to generate some anxiety, even in fully cooperative experiences, and the reluctance in cases of passive consent also is assumed to be manifested by anxiety, even when the child is knowledgeable about sex. Of particular interest are the two cells in Table 17-4 representing the "best case" and "worst case" conditions. The "worst case" situation, characterized by high anxiety and high guilt, is that of the "passive consenter" who is aware of moral negatives but ignorant of sex. This is the typical American child, told that sex is bad, but not what it is, and taught to be compliant and obedient with adults and older siblings and not to tattle. The "best case" situation is that of the active cooperator who knows about sex but has not absorbed conventional moral negatives

The American commune child described by Johnson and Deisher (1973) and children of the Muria tribe (Currier, 1979, and Chapter 2 this volume) would tend to fall into this category.

Other Factors

Such situational factors as the feelings communicated by the partner and the family's reaction to disclosure of the sexual incident can be presumed to affect a child’s degree of anxiety and guilt during or as a result of a sexual encounter with an adult.

The quality of family relations is of paramount importance in outcome. Negative reactions of parents (and other important adults) to a child's sexual encounters, aside from their function in inducing guilt, can be, at least in a minority of cases, the most psychonoxious aspect of the entire experience. When communication in a family is good and the family functions well, however, and especially when at least one parent can be highly supportive, even the effects of brutal assault may be minimized. Most incest occurs in otherwise seriously disturbs family settings; it appears that it is not the incest but the preexisting disturbed family relationships that are crucial in outcome.

Differentiating Sexual Experiences of Adults and Children

lt should be evident by this point that the factors found to influence reactions and outcome in childhood sexual experiences are not specific to childhood or limited to experiences prior to puberty or to attaining majority. Childhood sexual experiences do not differ qualitatively in their discernible effects from those of adulthood. If there are differences, they are of degree, not kind – of probabilities, not categoric certainties. The same factors are found to be operative in adult-adult sexual encounters, although our awareness of them may depend on special circumstances. A few examples make clear the parallels.

The issue of passive consent has assumed importance in recent discussions of sexual relations between adult college students and their professors. Often the student is apparently a willing participant but in fact experiences the situation as one that cannot be declined. Intense feelings of guilt, of debasement for having prostituted oneself, and of failure for not having resisted are reported, as are suppressed feeling of helpless rage toward the teacher and toward others without sympathy for the student.

The importance of a supportive family becomes clear in cases of rape. Adult rape victims also are sometimes blamed for their own victimization and accused of inviting the assault. Force and violent clearly take their toll here, creating a stark contrast not only to consensual adult sex but also to instances of passive consent.

Knowledge of sexual matters is not universal among adults, though adults are often presumed to know the "facts of life," such presumption must be questioned in some adult-adult sexual encounters, even as it is in child-adult sexual situations. Retarded and institutionalized adults, of course, are usually assumed to have lessened capacity for judgment, as well as less actual knowledge of sex (whether or not this is true of individuals).

In sum, the effects of sexual experiences during childhood, like the capacity for sexual reponse – like sexual awareness itself – are part of a continuum that clearly and systematically connects human sexuality from infancy through old age.

REMARKS

It is often simply assumed a priori that incest and sexual encounter with older persons are necessarily harmful to a child. In the substance literature on sexual abuse, sexual exploitation, and sexual molestation of children, few authors have even addressed the problem of definition, perhaps recognizing that any nontrivial definition of sexual abuse of children automatically defines, by exclusion, another category of sexual encounters that are not abusive. Brant and Tisza (1977) define sexual misuse, the term they prefer, in terms of "sexual stimulation appropriate for the child's age, ... psychosocial development, and role in the family." They further establish the criteria for inappropriateness as "symptoms in the child and evidence of family dysfunction." A similar definition is offered by Schechter and Roberge (1976), who favor the term sexual exploitation, involving "dependent, developmentally immature children and adolescents in sexual activities they do not fully comprehend, are unable to give informed consent to, and that violate the social taboos of family roles."

Both of these definitions, but especially the latter, are completely compatible with the conclusions of this review. Especially notable are the references to comprehension, consent, and family values. It should be clear that each of these definitions contains, by implication, a definition of healthy sexual encounters with children. Looked at this way, legitimate sexual experiences are therefore ones (1) in which the child is sexually knowledgeable and fully comprehends the activity; (2) to which he or she freely consents on the basis of that comprehension; (3) that take place in a family and/or social setting that affirms such sexual experiences as appropriate, and (4) that (therefore) do not result in symptoms of dysfunction in the child or the family.

A concept of legitimate instances of child-adult sex is supported both empirical evidence and reasoned definition. But this logical implication uncomfortably confronts the values and fears of many individuals in our society. It is precisely this fear of the unknown obverse to a familiar coin that has led to so much controversy and has hindered inquiry into the nature and effects of childhood sexual experiences.

REFERENCES

B